Air

Air pollution, particularly fine particulate matter (PM_2.5), poses significant public health and environmental risks. This study explores the effectiveness of spatiotemporal graph neural networks (ST-GNNs) in forecasting PM_2.5 concentrations by integrating remote-sensing hyperspectral indices with traditional meteorological and pollutant data. The model was evaluated using data from Switzerland and the Gauteng province in South Africa, with datasets spanning from January 2016 to December 2021. Key performance metrics, including root mean squared error (RMSE), mean absolute error (MAE), probability of detection (POD), critical success index (CSI), and false alarm rate (FAR), were employed to assess model accuracy. For Switzerland, the integration of spectral indices improved RMSE from 1.4660 to 1.4591, MAE from 1.1147 to 1.1053, CSI from 0.8345 to 0.8387, POD from 0.8961 to 0.8972, and reduced FAR from 0.0760 to 0.0719. In Gauteng, RMSE decreased from 6.3486 to 6.2319, MAE from 4.4891 to 4.4066, CSI from 0.9555 to 0.9560, and POD from 0.9699 to 0.9732, while FAR slightly increased from 0.0154 to 0.0181. Error analysis revealed that while the initial one-day ahead forecast without spectral indices had a marginally lower error, the dataset with spectral indices outperformed from the two-day ahead mark onwards. The error for Swiss monitoring stations stabilized over longer prediction lengths, indicating the robustness of the spectral indices for extended forecasts. The study faced limitations, including the exclusion of the Planetary Boundary Layer (PBL) height and K-index, lack of terrain data for South Africa, and significant missing data in remote sensing indices. Despite these challenges, the results demonstrate that ST-GNNs, enhanced with hyperspectral data, provide a more accurate and reliable tool for PM_2.5 forecasting. Future work will focus on expanding the dataset to include additional regions and further refining the model by incorporating additional environmental variables. This approach holds promise for improving air quality management and mitigating health risks associated with air pollution.

Road dust resuspension is widely recognized as a major contributor to traffic-related particulate matter (PM) in urban environments. Nevertheless, reported emission factors exhibit substantial variability. These discrepancies stem not only from the intrinsic complexity of the resuspension process but also from limitations in measurement techniques, which often fail to adequately control or characterize the influencing parameters. As a result, the contribution of each parameter remains difficult to isolate, leading to inconsistencies across studies. This study presents an experimental protocol developed to quantify PM10 and PM2.5 emission factors associated with vehicle-induced road dust resuspension. Experiments were conducted on a dedicated test track seeded with alumina particles of controlled mass and size distribution to simulate road dust. A network of microsensors was strategically deployed at multiple upwind and downwind locations to continuously monitor particle concentration variations during vehicle passages. Emission factors were derived through time integration of the mass flow rate of resuspended dust measured by the sensor network. The estimated PM10 emission factor showed excellent agreement, within 2.5%, with predictions from a literature-based formulation, thereby validating the accuracy and external relevance of the proposed protocol. In contrast, comparisons with U.S. EPA formulas and other empirical equations revealed substantially larger discrepancies, particularly for PM2.5, highlighting the persistent limitations of current modeling approaches.

Fine particulate matter (PM_2.5) poses substantial urban health risks that vary across space, time, and population vulnerability. We integrate a spatio-temporal INLA–SPDE PM_2.5 field with an agent-based model (ABM) of 10,000 daily home–work commuters in Indianapolis’s Pleasant Run airshed (50 weeks; 250 m grid). The PM_2.5 surface fuses 23 corrected PurpleAir PA-II-SD sensors with meteorology, land use, road proximity, and MODIS AOD. Validation indicated strong agreement (leave-one-out R² = 0.79, RMSE = 3.5 μg/m³; EPA monitor comparison R² = 0.81, RMSE = 3.1 μg/m³). We model a spatial-equity counterfactual by assigning susceptibility independently of residence and workplace, isolating vulnerability from residential segregation. Under this design, annual PM_2.5 exposure was statistically indistinguishable across groups (16.22–16.29 μg/m³; max difference 0.07 μg/m³, <0.5%), yet VWDI differed by ~10× (High vs. Very Low). Route-level maps reveal recurrent micro-corridors (>20 μg/m³) near industrial zones and arterials that increase within-group variability without creating between-group exposure gaps. These findings quantify a policy-relevant “floor effect” in environmental justice: even with perfect spatial equity, substantial health disparities remain driven by susceptibility. Effective mitigation, therefore, requires dual strategies—place-based emissions and mobility interventions to reduce exposure for all, paired with vulnerability-targeted health supports (screening, access to care, indoor air quality) to address irreducible risk. The data and code framework provides a reproducible baseline against which real-world segregation and mobility constraints can be assessed in future, stratified scenarios.